I have two interesecting geopandas.GeoDataFrame datasets, one containing line features (lines_gdf) and one containing polygons (polygons_gdf):

import matplotlib.pyplot as plt
import geopandas as gpd
from shapely.geometry import Polygon, LineString

# Create sample polygon and line data
polygons_gdf = gpd.GeoDataFrame(geometry=[Polygon(zip([150, 151, 151, 150], 
                                                      [30, 30, 31, 31]))])
lines_gdf = gpd.GeoDataFrame(geometry=[LineString(zip([148, 150, 152], 
                                                      [28, 30, 32]))])

# Plot example data
lines_gdf.plot(ax=plt.gca(), edgecolor='red')

Example of intersecting line and polygon datasets

Desired output

I want to use my polygons dataset to remove or erase any part of my lines dataset it intersects with (essentially a GIS "erase" operation), and return only the non-overlapping parts of my lines (i.e. black lines below):

enter image description here

Using the geopandas.clip tool I've been able to do the exact opposite of what I need to do: clip the data to keep (not remove) the intersecting portion of my lines:

# Clip lines to keep only area within `polygons_gdf`
clipped_gdf = gpd.clip(lines_gdf, polygons_gdf)

# Plot clipped lines over the top of input layers
lines_gdf.plot(ax=plt.gca(), edgecolor='red', alpha=0.1)
clipped_gdf.plot(ax=plt.gca(), edgecolor='black')

enter image description here

My question: How do I use geopandas to remove/erase part of a line dataset that intersects with a polygon dataset?

My geopandas version is v0.6.3.

2 Answers 2


A colleague helped me out with this one:

  1. Upgrade geopandas to version 0.7.0 or above. Prior to version 0.7.0, this function supported polygon operations only. Version 0.7.0 introduced the following change:

The geopandas.overlay function now works for all geometry types, including points and linestrings in addition to polygons

  1. Use the geopandas.overlay function with how='difference':
erased_gdf = gpd.overlay(lines_gdf, polygons_gdf, how='difference')

This now gives me the result I required:

# Plot output over the top of input layers
lines_gdf.plot(ax=plt.gca(), edgecolor='red', alpha=0.1)
erased_gdf.plot(ax=plt.gca(), edgecolor='black')

enter image description here


I haven't used geopandas much, but when I did, I needed a similar operation. Try looking at the set operations. The one you're after is "symmetric difference".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.